Update app.py

app.py

@@ -1,50 +1,30 @@
-import gradio as gr
-import spacy
-from kommunicate import Kommunicate
+from flask import Flask, request, render_template
+from transformers import pipeline
 
-# Load the Spanish model for spaCy
-nlp = spacy.load('es_core_news_sm')
+app = Flask(__name__)
 
-# Map the Spanish POS tags to their English equivalents
-pos_map = {
-    'sustantivo': 'NOUN',
-    'verbo': 'VERB',
-    'adjetivo': 'ADJ',
-    'artículo': 'DET'
-}
+nlp = pipeline('sentiment-analysis')
 
-# Function to identify the POS tags in a sentence
-def identify_pos(sentence):
-    doc = nlp(sentence)
-    pos_tags = [(token.text, token.pos_) for token in doc]
-    return pos_tags
+@app.route('/')
+def home():
+    return render_template('index.html')
 
-# Game logic
-def game_logic(sentence, user_word, user_pos):
-    correct_answers = identify_pos(sentence)
-    user_pos = pos_map[user_pos.lower()]
-    for word, pos in correct_answers:
-        if word == user_word:
-            if pos.lower() == user_pos.lower():
-                return True, f'¡Correcto! "{user_word}" es un {user_pos}.'
-            else:
-                return False, f'Incorrecto. "{user_word}" no es un {user_pos}, es un {pos}.'
-    return False, f'La palabra "{user_word}" no se encuentra en la frase.'
+@app.route('/predict',methods=['POST'])
+def predict():
+    if request.method == 'POST':
+        message = request.form['message']
+        prediction = nlp(message)
+        return render_template('index.html', prediction_text=prediction)
 
-# Main function for the Gradio interface
-def main(sentence, user_word, user_pos):
-    if sentence and user_word and user_pos and user_pos != 'Selecciona una función gramatical...':
-        correct, message = game_logic(sentence, user_word, user_pos)
-        return message
-    else:
-        return 'Por favor, introduce una frase, una palabra y selecciona una función gramatical válida (sustantivo, verbo, adjetivo, artículo).'
+if __name__ == "__main__":
+    app.run(debug=True)
+from transformers import GPT3LMHeadModel, GPT2Tokenizer
 
-# Create the Gradio interface
-iface = gr.Interface(fn=main, 
-                  inputs=[
-                      gr.inputs.Textbox(lines=2, placeholder='Introduce una frase aquí...'),
-                      gr.inputs.Textbox(lines=1, placeholder='Introduce una palabra aquí...'),
-                      gr.inputs.Dropdown(choices=['Selecciona una función gramatical...', 'sustantivo', 'verbo', 'adjetivo', 'artículo'])
-                  ], 
-                  outputs=gr.outputs.Textbox())
-iface.launch()
+tokenizer = GPT2Tokenizer.from_pretrained("gpt2")
+model = GPT3LMHeadModel.from_pretrained("gpt3")
+
+def get_response(prompt):
+    inputs = tokenizer.encode(prompt, return_tensors="pt")
+    outputs = model.generate(inputs, max_length=150, num_return_sequences=1, no_repeat_ngram_size=2)
+    response = tokenizer.decode(outputs[:, inputs.shape[-1]:][0], skip_special_tokens=True)
+    return response